I have Illumina RNA-Seq data with extremely short fragment sizes (150bp for 100bpPE reads). I am using Trimmomatic because I like how the palindrome clipping handles read-through (and bonus points for handling orphans for me). However, after running Trimmomatic I still see residual adapter contamination in the last 7bp or less of the 3' end of some of the reads.
Looking at the source code, I see there is a parameter MIN_PREFIX=8 (in IlluminaClippingTrimmer.java). When I change it to MIN_PREFIX=1, I get the desired result. I think I understand the code, but I don't code in Java. Is there a reason that I am missing that makes it a bad idea to lower this value? My reasoning is that if the left and right reads overlap in such a way that indicates read-through, even if only 1 base of the adapter was sequenced, then we can safely trim the small bit of adapter and throw out the second read.
Looking at the source code, I see there is a parameter MIN_PREFIX=8 (in IlluminaClippingTrimmer.java). When I change it to MIN_PREFIX=1, I get the desired result. I think I understand the code, but I don't code in Java. Is there a reason that I am missing that makes it a bad idea to lower this value? My reasoning is that if the left and right reads overlap in such a way that indicates read-through, even if only 1 base of the adapter was sequenced, then we can safely trim the small bit of adapter and throw out the second read.